【正文】 ied out laboratory tests that measured uplift loads acting on horizontal plate, ??type deck with downstanding longitudinal beams in shallow water, and horizontal plate exposed to breaking waves, and then proposed an empirical model. Zhou and Chen ( 2003, 2004) reported another experimental study on wave uplift on horizontal slab, taking into account the effect of the clearance, the wave steepness and the width of deck. The wave uplift is general ly considered to be made up of two ponents, one is short duration impact force and another is long duration force. The impact ponent mostly exceeds the slowvarying one and in some cases up to more than 10 times in magnitude. Owing to the plexity of influencing factors related to the impulsive uplift load, guidance with respect to the force is not readily available. Most of the existing models are presented based on regular wave tests, and the measured data is very dispersed over the range of measurements for above methods, which results in significant discrepancies among the results of different models, even though the existing methods can likely provide adequate estimates of loads under their own situations. Further research is strongly remended for systematic investigation.2. Experimental Setup and ProcedureThe wave tests were conducted in a wave flume of 1. 0 m wide, 1. 2 m deep, 80 m long, as shown in Fig. 1. The service part of the flume was divided into two parts, 0. 5 m and 0. 5 m respectively. One is the test section。 mean incident wave period Tm= 1. 0, 1. 2, 1. 5, 2. 0, 2. 5 s) . The width of deck was B = 30, 50, 80, 102 cm. Six different ratios of clearance ??h to the significant incident wave height ??h/H s were tested: ??h/H s= 0, 0. 1, 0. 2, 0. 3, 0. 4, 0. 6. Water depth at the jetty model was d= 50 cm. The interval wave generation form was used to dissipate the multiple reflections of waves. The duration for each wave set was 5~ 9 min and the wave number was 120~ 150. Each group of tests was repeated 3 times so that the reliability of the measured data could be guaranteed. The sampling time interval of data acquisition was 1/ 125 s. About 200 sets of tests were conducted under the conditions mentioned above. Fig. 1. Sketch of the experimental setup of the exposed high pile jetty ( unit: cm, scale: 1: 36) .3. Results and Discussions3. 1 Distribution Length Associated with Uplift Force on HighPile Jetty DeckThe spatial pressures distribution associated with uplift forces have been categorized as the impulsive type and the uniform type from tests shown in Fig. 2~ Fig. 4. The maximum uplift loads generally lag behind the maximum impulsive pressure and are associated with the pressure uniformly distributed. It means that the uplift force related to the maximum impulsive pressure is not the largest. For uniform distribution, the corresponding pressure is relatively small, while the distribution length is large. The distribution length increases with the wave length and clearance decrement . On the contrary, with regard to the impulsive type distribution, the pressure is extremely large but localized in a small area. The relevant distribution length also conforms to the trends as mentioned above. The estimation of the distribution length varies among different researchers and methods. For example, Guo and Cai (1980) uses L / 4, while Guoda uses L / 9~ L / 6. Selecting a constant ratio of wave length here to denote the distribution length is unreasonable because the values for cases with identical wave length would remain constant even for different clearances, which is clearly inconsistent with the phenomenon recognized in experiment that the distribution length corresponding to large clearance is small. It would result in overestimation of the uplift loads. It could be understood that the uplift pressure acting surface ( namely the wave acting surface) is strongly linked to the wave contact length. Therefore, taking the wave contact length to express the distribution length seems more reasonable. Generalization and analysis of the experimental data confirmed that the distribution length of the uniform type was equivalent to x . When x is larger than the width of deck B, it is taken as B. Fig. 2. Simultaneous pressure distribution on the deck ( uniform type, the corresponding wave contact length x 1% = 0. 93 m) . Fig. 3. Simultaneous pressure distribution on the deck ( uniform type, the corresponding wave contact length x 1% = 0. 5 m) . Fig. 4. Simultaneous pressure distribution on the deck ( impulsive type, the corresponding wave contact length x = 0. 93 m) .3. 2 WaveinDeck Uplift Loads on the Exposed HighPile Jetty3. 2. 1 Parametric Analysis of the Eff ect of Geometric and Hydrodynamic VariablesAnalysis of the measured data confirms that the dominant variables influencing the uplift force are wave height, wave length, the clearance of deck, and the width of deck.3. 2. 1. 1 Effect of clearanceDimensionless uplift load on deck . the relative clearance is plotted in Fig. 5.where P 1% denotes the maximum value of the waveindeck uplift force per unit length of deck ( with exceedance probability 1% ) , and the direction of that length is perpendicular to the wave propagation direction。 ??h is deck clearance。 L s is the significant incident wave length ( associated with significant wave period) 。 coefficient 1. 1 is the amplification coefficient attributed to the wave reflection. When x 1% is larger than the width of deck B , it is taken as B。 the point of peak pressure is shifting up and the width of the area experiencing wave action increases. The effect of the wave dynamics on the impulsive pressure enhances with the clearance increment. At a low clearance level, the wave height has little effect on the impulsive pressure in respect that the air cushion plays a role in mitigating and uniformizing pressure. However, with regard to a high clearance level, there is no air cushio